turn ons

I get my kicks from reading poems and beautifully constructed theories. They're both loaded with such tremendous meaning and so much promise.That's what attracts me to them.

I love reading mathematical theories and proofs. I never fully understand what they're about and I can never grasp them 100%. But I love reading about century-old mathematical puzzles being solved, elegant proofs laid out for complicated problems, complicated proofs to solve seemingly simple problems...

Recently, an Indian mathematician was in the limelight for winning the coveted Abel Prize ( == the Nobel Prize in Mathematics). The following was a conversation I had with a colleague from work. I am yet to figure out most of his explanation, but I was still totally turned on by this conversation.. (yes, honestly)

its about some special type of markovian/stochastic processes..
for example you know about central limit theorem .. right?
that a distribution becomes Normal when there are enough population..
me: yeah yeah i remember that one
right.. so this theorem tells us that eventually a independent and identical distribution will become a normal distribution..
that is... if we compute the distribution on ENOUGH number of samples then we will get a normal distribution. but it was not known HOW MUCH is ENOUGH
in other words.. suppose we have a sample.. now we want to know how close we are to the true normal distribution.
.. OR ..
How fast we can reach the true distribution if we go on collecting samples
... Large Deviation answers these questions.
so that we can practically and effectively know how much data we need, how much is sufficient, how far we are from true distribution.. ..and how to detect rare events... in case we get erratic distributions as we include more samples
i.e., we don't see any particular convergence. Now that can be interesting in detecting fraud..or validating traffic ...there are hardly any mechanisms known till date to detect rare events of large data just by considering a small sample Large Deviation can provide a solution to this...and if it does... traffic estimation and computations will be highly real time and effective...

And reading poetry is my other fascination .
There's T. S. Eliot
"Let us go then you and I ,
when the evening is spread out against the sky,
like a patient etherised upon a table"

and Neruda
"...Mother of stone , spume of the condors.."

and Dylan Thomas
" Rage , rage against the dying of the light.."

and so many more brilliant pieces.


I wish I could spend days just reading such brilliance, being repeatedly fascinated and turned on (for lack of a better word).